fmemuir / coastguard Goto Github PK

Bug appears when adjust_detection() is called to adjust the NDVI threshold used for veg edge extraction. If a point is clicked, the plot updates with the new veg edge position relating to that updated NDVI threshold. If no point is clicked however, and the user hits 'Enter' without clicking anything, the plot breaks with the error:
vegline_latlon referenced before assignment

Investigating to see where vegline might have been defined but not vegline_latlon or vegline_proj

Hi - Excited to get started with VedgeSat and COASTGUARD to compliment some CoastSat data!

I've been trying to work through the example scripts and am having difficulty with the "check_images_available" function. Both the VedgeSat_DriverTemplate.ipynb and VedgeSat_Driver_Example.py file create the same issue for me where the script becomes nonresponsive and does not update when checking for available images. This issue persists regardless of area or timeframe and seems like it is stuck in an infinite loop near line 420. Any thoughts?

ResolvePackageNotFound errors on packages detailed in attached text file.

COASTGUARD_environment_errors.txt

Also, the environment .yml filename needs to be renamed with the new COASTGUARD branding.

Hi Freya et al.

I've been working a bit with the VedgeSat toolbox on a study site in NE Florida USA using Landsat and Sentinel 2 data and using the provided example driver notebooks. While the vegetation and water level detections appear to be working as shown in the attached classification image, the output .pkl files are empty (screenshot). Any thoughts on how I can address this?

73723280/56e04629-ff18-44ac-83b7-4561bae52cef)

Code syntax error in Image Download cell of Jupyter Notebook driver. Attempting to resolve myself and if I can I will post solution here.

In certain sites and images, the veg edge extraction process falls over when contours are attempted to be extracted. Specifically at:

File COASTGUARD/VegetationLine.py, line 232
    contours_ndvi, t_ndvi = FindShoreContours_WP(im_ndvi, im_labels, cloud_mask, im_ref_buffer)
File COASTGUARD/VegetationLine.py, line 1030
    t_ndi, _ = Toolbox.FindWPThresh(int_veg, int_nonveg)
File COASTGUARD/Toolbox.py, line 2013
    model.fit(sample)

The code fails because model cannot take a variable of shape (0,1) (i.e. sample is empty). Tracing this back, this happens on very hazy/dark/poor quality images, which have somehow passed the initial catches for image quality (cloud masking, empty rasters). The reason the array going into FindWPThresh is empty is because not enough classes have been found in order to define a contour along; most likely all pixels have been classes as nonveg.

This needs to be fixed by either catching these images early and discarding based on reflectance histograms, or the slightly faster route of a conditional after the images have been classified which skips this image if the labels are all one class.

Might be to do with the reference line shapefile itself, but certain sites fail to create a proper reference line buffer that runs along the reference line shapefile. The shapefile itself looks good (all one feature, no breaks or multiparts), but when the buffer is generated, it only covers the first 20% of the line. One temporary fix for this is to expand the minimum distance for the buffer so that it covers the whole reference shoreline. But it should be investigated to make sure something weird isn't happening with Toolbox.ProcessRefline() which is causing the buffer to be cut off early. Perhaps a reprojection bug? Or a bug in Toolbox.spaced_vertices()?

Hi @fmemuir,

Running into issues with the operation of the new RemoveDuplicates function (approximately Line 600 in Toolbox.py).

It is a list out of range error which I think stems from the fact that the output['veglines'] list reduces in size as duplicates are deleted. The additional index 'v' appears to refer to values outside of the range of the length of this list, likely ObjectIDs of some sort?

Hi, I was trying to run the VegetationLine.extract_veglines but it encountered an error related to geemap.ee_to_numpy function in the Image_Processing.py module (line 523) -> TypeError: ee_to_numpy() got an unexpected keyword argument 'scale'.

I read the geemap documentation related to ee_to_numpy() and the parameter is set to None but you can pass it.

If you have any idea about it, I would appreciate it.

Thanks!

Hi Freya,

I encountered a problem when I attempted to extract the vegetation edge in year 2024 using Sentinel-2 images. I received the error message "Image does not contain the band QA60". However, when I change the date to 2023, it works as expected.